1
Achieving the Triple Aim for Children with
Medical Complexity
Session 302, February 15
th
, 2019
Christopher J. Stille, MD, MPH, Sara E. Martin RN, BSN
Children’s Hospital Colorado
2
Christopher J. Stille, MD, MPH and Sara E. Martin RN, BSN
have no real or apparent conflicts of interest to report.
Conflict of Interest
3
• Discuss our organization’s experience implementing
technology solutions during a multi-hospital learning
collaborative aimed at improving care coordination for
medically complex children
• Review implementation process of multiple care
coordination tools integrated within our electronic
medical record
• Analyze patient outcomes seen after a hospital wide
implementation of integrated complexity scoring and
care coordination tools
Agenda
4
At the end of this session, learners will be able to:
• Describe tools and technology to improve patient and family
satisfaction with the care they receive
• Explain the benefits of combining social complexity with medical
complexity when determining and assigning patient risk
• Discuss how IT tools can improve outcomes and decrease costs
for medically complex patients
Learning Objectives
5
An entire hospital,
devoted to kids
Numbers from 2017
Kids aren’t
small adults
They’re growing and changing every
day, and their care not only has to
address what they need right now;
it also has to anticipate where
they’ll be in 6 months, 9 months, 10
years. That’s pediatric expertise.
We treat kids and only kids. Our
physicians are board-certified,
meaning they exceed state licensing
requirements and commit to
continually expanding their
knowledge in their specialty area.
Our nurses are specially trained in
pediatrics and hold Bachelor of
Science degrees in nursing.
Our expertise shows in our results:
each kid we treat, soothe and heal,
each kid who gets well enough to
walk out our doors. It’s why we do
what we do.
479
Licensed beds
173,085
Emergency and urgent
care visits
15,330
Inpatient admissions
6,821
Employees
19,362
Total surgeries
2,170
Medical staff
97,418
Days of patient care
259
Residents and fellows
574,929
Outpatient visits
2,502
Volunteers
6
Coordinating All Resources Effectively (CARE)
A Multi-center Learning Collaborative
Project Goals:
• Improved health outcomes for kids with Medical Complexity
• Improved family satisfaction with access to care
• Decreased cost of care
•
Medicaid Medical Complexity Partnership
CMMI Health Care
Innovation Award
(Centers for Medicare
and Medicaid Innovation)
3yr Learning Collaborative
10 partner Hospitals
$23 Million Award
CRG Groups 5B-9
(Clinical Risk Grouper)
7
Medicaid
National Level
Children’s Hospital Colorado
# of children % of children
percent of
spend
# of
Children
% of Children
Total Spend
Healthy/ episodic/
chronic (CRG 0
-5a)
31
million
94% 60% 596,700 94% Unknown
Medically complex
(CRG 5b
-9)
2 million
6% 40% 38,300 6%
$649.7M
CARE Enrolled
Population
--- --- --- ~ 1600
0.22%
$73.6M
Total
33
million
100% 100%
635,000
100% Unknown
Children With Medical Complexity (CMC)
in Colorado
Note: “Percent of spend” includes <1 year olds with < 6
months continuous enrollment.
Source: Extrapolated from Truven Marketscan
Medicaid claims dataset, 2011.
8
Governance structure
Care Coordination Steering
(Executive Leadership)
CARE Award Team
Physician Leaders
Grant Funded Internal Staff
Community
PCP Partners
Health Literacy
Project Management
Process Improvement
Process Improvement
Research Data Analyst
Local PCP’s
RN Care Coordinators
Family Navigators
Care Coordinator
Case Management
CHCO Social Work
Primary Care leadership
EHR Analysts
Social Worker / Family
Liaison
Behavioral Health
9
9
The Change
Package
Patient Registry
• All enrolled children
• Integrated within EHR
• Accessible and transparent
Dynamic Care Team
• Accessible by parents and PCP
• Integrated
• Easily revised
Access Plan
• Who and when to call for what symptoms
• Contact info for all team members.
Shared Care Plan
• Patient and family’s specific needs
• Short and long term goals
• “Who am I??”
Goals
• Integrated
• Accessible
• Transparent
10
10
Care Coordinator:
“I can’t find
important
information”
Care Coordinator:
“I don’t know who to
focus on first”.
Parent:
“I’m doing Care
Coordination for
my own child”
Voice of the
Customer
Community PCP:
“ We don’t know
who to call when
we need help”.
Parent:
“I never know
who knows what
about my child”
11
Family Experience
• CAHPS- Patient Access Module
(Consumer Assessment of Healthcare Providers and Systems)
Utilization
• ED visits
• Hospitalization (Inpatient days)
Direct Cost
• Total cost of care
Measure n
Baseline
Jan-Dec 2014
CAHPS
Patient Access Module Score
N=111
69.3
IP Days
1000pts/month
N=1467
297 days
ED visits
1000pts/month
N=1467
80 visits
Total Direct Cost
(aggregate population)
N=1481
$23,405,724
($15,804 per pt)
Note: N number for CAHPS reflects number of enrolled families able to be reached by
phone to successfully complete both pre and post survey.
Baseline Data
12
1 2 3 4 5
Integration
within the EHR
Define their
needs
Define / Draft
Technology
requirements
Define our
population
Prioritize and
deploy care
Our Approach
• Complexity
Score
• Patient Registry
• Develop
complexity
algorithm
• Kaizen events
• HIT Integration
• Transparency
• Medical / Social
Complexity
• “One place”
• Clinical tool
development
• Streamlined
processes
• Tools into
Practice
WHAT:
HOW:
13
Population and Needs Defined
• Three day Kaizen
• 4 Tier system
• Medical and social
complexity
• Interventions escalate
with complexity
• Change package for
tiers 3 and 4
14
Medical Complexity:
Severity Adjusted Risk Categories
Complexity
increases as
Risk Category
increases
15
Psychosocial Screen:
Social Needs Domains
Finding a Medical Home
Barriers to Keeping Appointments
Barriers to Purchasing Medications
Financial Resource Strain
Food Insecurity
Access to Benefits
Education
Housing Instability
Guardianship
Social Isolation
Substance Abuse in Caregiver
Family Violence
Depression/Anxiety in Caregiver
Social Complexity:
Incorporating Social
Determinants of Health
Standard Screening Tools Included
•
ASQ (Ask Suicide-Screening Questions)
•
CRAFFT
(Screening tool for Adolescent Substance Abuse)
•
PHQ-9 (Patient Health Questionnaire-9)
•
EPDS (Edinburgh Postnatal Depression Scale
• 13 domain Psychosocial Screen
• Standardized screens also incorporated
• Complexity increases for positive screens
16
Social Determinants of Health Assessment:
Clinical Process
17
Social Determinants:
Clinical Decision Support
• Clinical decision support tools
integrated into clinical workflows
• Clinicians notified real time of
positive screens needing follow up
• All data reportable and tracked
18
From Imagined to Integrated:
• Wellness Registry: store and
track data and outcomes
• Allows data sharing with
PCP’s, schools
• All children seen last 3 years
• CRG, social determinants,
homecare and ED visits
• Stores complexity score over
time
• Tier 3 and 4 children
selected for change package
interventions ~1600 kids
19
Patient Complexity Scoring:
Ideas into Practice
• “Hover to discover”
• Real-time Updates
• All elements
integrated within EHR
20
Longitudinal Plan of Care
1
2
3
21
Longitudinal Plan of Care
Longitudinal Plan of Care
available below provider
schedule
1) Clinicians can choose
columns to add
2) Complexity Score
elements clearly visible
3) Psychosocial screen
accessible from report
1
2
3
22
Care Team,
Care Plan
and Access
Plan:
Integration
• Snapshot of “Who am I?”
• “Who to call for what”
• Patient / family specific
• Available to community providers,
homecare companies and school
through web portal
• Family accessible through patient
portal
23
23
Parent View: Longitudinal Plan of Care
View from the online
Parent Web portal
24
Results:
Measure n
Baseline
Jan-Dec 2014
n
Post
Implementation
Jan-Dec 2017
% change
CAHPS
Patient Access Module Score
n=111
69.3
n=111
Fall-Spring
2016-2017
81.0
↑16%
p<0.05
IP Days
1000pts/month
n=1467
297 days
n=1588
123 days
↓58.6%
(goal=40%)
ED visits
1000pts/month
n=1467
80 visits
n=1588
54 visits
↓32.5%
(goal=10%)
Direct Cost per
Adjusted Volumes/year
n=1481
$23,405,852
(
Aggregate population
)
n=1529
$14,221,683
↓39.4%
(goal=6.8%)
Note: Patients were enrolled during the first full year of the collaborative.
Pre-implementation N reflects # patients enrolled in June 2015. Post
implementation N reflects total number enrolled at end or collaborative,
accounting for new enrollees during first year and deaths.
25
-100
100
300
500
700
900
1100
1300
1500
1700
Change Package Implementation
# Enrolled Patients Care Team Access Plan Care Plan
Enrolled Patients
Process Measures:
Care Team, Access Plan, and
Care Plan
• Care team and Access
plan automated in EMR
• Care plan required 1:1
intervention with families
resulting in longer
implementation times.
26
26
Care Coordinator:
“All the important
information is in
one place”
Care Coordinator:
“I can easily see
my Tier 3 and 4
patients for today’s
clinic”.
Parent:
“I know exactly
who to call for
what”
Voice of the Customer
Community PCP:
“ The updated Care
team and access
plan have been a
huge help”.
Parent:
“The Care Plan is
easy to understand, I
know exactly where
to go”
27
69.3
81.0
0.0
20.0
40.0
60.0
80.0
100.0
Patient Access Module Score,
Converted to Percentage
Baseline
Final
Whole Collaborative
CAHPS Survey: Access
Scores
Children’s Colorado ONLY site
with a statistically significant
improvement!
Patient Access: Results by Site
1 2 3 4 5 6 7 8 9 10
Site number
28
0
20
40
60
80
100
120
Jan-14
Apr-14
Jul-14
Oct-14
Jan-15
Apr-15
Jul-15
Oct-15
Jan-16
Apr-16
Jul-16
Oct-16
Jan-17
Apr-17
Jul-17
Oct-17
Jan-18
Apr-18
Jul-18
Oct-18
Emergency Department Visits
per/1000 patients/month
0
50
100
150
200
250
300
350
400
Jan-14
Apr-14
Jul-14
Oct-14
Jan-15
Apr-15
Jul-15
Oct-15
Jan-16
Apr-16
Jul-16
Oct-16
Jan-17
Apr-17
Jul-17
Oct-17
Jan-18
Apr-18
Jul-18
Oct-18
Inpatient Days
per/1000 patients/month
Collaborative
Concludes
Interventions
Begun
Enrollment
Begins
Utilization
2014-2018
Collaborative
Concludes
Interventions
Begun
Enrollment
Begins
29
Relationship:
ED visits and Change Package elements
29
As more change
package elements
are implemented,
ED rate drops.
% of Change package implemented
Note: Graphic obtained from Children’s Hospital Association
30
Direct Costs: Did we change anything?
↓ 55%
cost of Inpatient care
per enrolled patient
(2015-2017)
↑ 7.1%
cost of Ambulatory Care
per enrolled patient
(2015-2017,ED costs are included in
ambulatory setting)
↓ 33%
total direct cost per patient
per year
(2015-2017)
$8,881
$5,007
$4,009
$4,940
$4,822
$5,292
$0
$2,000
$4,000
$6,000
$8,000
$10,000
2015 2016 2017
Direct Cost per Unique Patient
Direct Cost per Unique Patient by Patient Type
Inpatient Outpatient
31
Cost Assessment
Eliminating the age progression factor
$-
$10,000
$20,000
$30,000
$40,000
$50,000
2014 2015 2016 2017
Enrolled Tier 4 patients:
Total Direct Cost
per patient per yr
With Careplan Without Care Plan
Linear (With Careplan) Linear (Without Care Plan)
• Due to resources, not all patients
were able to receive a care plan
during the 3 year project
• Tier 4 patients that did receive a
care plan appeared to have a
faster decline in costs than those
that did not receive a care plan.
• More data should be evaluated
on larger population to determine
if trend continues with expansion
• Graph represents 391 Tier 4 patients enrolled in the collaborative
• Blue line: 125 tier 4 patients with a completed care plan by
end of collaborative
• Red line: 266 tier 4 patients without completed care plans by
end of collaborative.
32
32
Lessons Learned:
Social determinants of
health integration critical
to successful tiering of
medically complex kids
Get interdisciplinary
team and parent
feedback frequently
during build process
Technical build
requires multiple
iterations...
Validate often!
Designing effective care coordination solutions for children
with medical complexity is very difficult...
But not impossible
33
Conclusions
• IP hospital days, ED visits, and
total direct cost were
decreased in a population of
medically complex kids after
targeted care coordination
interventions
• Satisfaction increased for both
families and providers
34
Moving Forward
• Expand implementation in all
ambulatory clinics
• Expand Social Determinants to
community practices
• Collaborate with local health
systems for shared care across
the continuum
• Share EHR based solutions with
other children's health systems
across the country
35
Christopher J. Stille, MD, MPH
Christopher.Stille@childrenscolorado.org
Sara E. Martin RN, BSN
Sara.Martin@childrenscolorado.org
Please remember to complete your session evaluation!
Questions